New Zealand Journal of Marine and Freshwater Research, 1986, Vol. 20 : 439-445
0028-8330/86/2003-0439$2.50/0
© Crown copyright 1986
439
Isolation of anaerobic, extremely thermophilic,
sulphur metabolising archaebacteria
from New Zealand hot springs
B. K. C. PATEL'
P. M. JASPERSE-HERST
H. W. MORGAN
R. M. DANIEL
Department of Biological Sciences
University of Waikato
Private Bag, Hamilton
New Zealand
INTRODUCTION
Interest in the microbiology of extreme environments, especially the geothermal environment from
which extremely thermophilic anaerobic archaebacteria have been isolated, has been increasing for
the past decade. Archaebacteria, which form the
third Urkingdom of organisms (Fox et al. 1980),
have provided a different perspective on the evolution of life and hence are of interest to evolutionary
biologists. The possibility that these organisms,
which have optimum growth temperatures between
Abstract
Enrichments of New Zealand geo- 80 and 105°C (Fischer et al. 1983; Stetter et al.
thermal samples, initiated in anaerobic sulphur- 1983), can be exploited to develop novel biotechcontaining media and incubated at temperatures nological processes, e.g., production of thermostaabove 85°C, yielded rod and coccal shaped organ- ble enzymes (Buonocore et al. 1980), has interested
isms which possessed archaebacterial characteris- industrial microbiologists and biotechnologists.
tics. Pure cultures were isolated and characterised.
A number of extremely thermophilic anaerobic
Five of the seven isolates, which were rod-shaped archaebacteria with growth optima between 85 and
organisms and did not have an obligate require- 100°C have been described. These include Therment for sulphur respiration, were similar to Ther- moproteus sp. (Zillig et al. 1981; Fischer et al. 1983),
moproteus sp. but had more neutral pH optima for Desulfurococcus sp. (Zillig et al. 1982), Thermogrowth. Three of these five Thermoproteus sp. were filum pendens (Zillig et al. 1983a), Pyrodictium sp.
obligate heterotrophs, which has not previously (Stetter et al. 1983), Thermococcus celer (Zillig et
been reported. The two coccal isolates had an obli- al. 1983b), and the methanogenic Methanothermus
gate requirement for sulphur as an electron accep- sp. (Stetter et al. 1981).
tor and were similar to Desulfurococcus sp. but
All of these organisms have been isolated from
again with more neutral pH optima for growth.
thermal areas of the Northern Hemisphere (Italy,
Iceland, Japan, and USA) and have not been
Keywords
hot springs; archaebacteria; Ther- reported from the thermal areas of the Southern
moproteus; Desulfurococcus; sulphur respiration; Hemisphere. Geography appears to be an important factor in their distribution since Thermoprogeothermal; growth; optima
teus sp. and Desulfurococcus sp. could not be
isolated from a number of springs in Italy and Japan
(Zillig et al. 1981, 1982). Pyrodictium sp. and Thermococcus celer were isolated from Italian but not
from Icelandic springs (Zillig et al. 1983b). Studies
of the thermal springs of the Southern Hemisphere,
especially those of New Zealand which have a
higher boiling point and generally higher sulphide
content than most thermal springs from other parts
of the world (Brock 1978), might reveal novel types
Received 28 November 1985; accepted 3 February 1986
of archaebacteria. We report here the ecology and
'Present address: Ruakura Animal Research Station, physiology of some archaebacteria found in New
Ministry of Agriculture and Fisheries, Private Bag,
Zealand thermal springs.
Hamilton, New Zealand.
New Zealand Journal of Marine and Freshwater Research, 1986, Vol. 20
440
Table 1 Results of enrichment of bacteria in PEM at an
incubation temperature of 92°C.
Pool
no.
Temp.
CQ
pH
70
64
82
97.5
82
96-102
97
100
97
82
90
97
103
103
97.7
94.5
95
98.2
98.5
97
NA
NA
91
NA
94
95
94
98
96
96
100
98
7.2
9.1
6.9
6.9
7.2
8.7
3.69
7.4
7.3
7.3
7.9
7.2
8.1
7.2
7.2
8.1
7.5
3.4
3.3
6.6
NA
NA
6.76
NA
7.8
8.3
8.2
7.3
7.6
6.8
8.5
4.8
85
88
96
80
81
79
67
3.03
3.1
8.5
8.5
7.34
7.7
6.5
97
3.15
96
6.5
89
6.7
97
6.9
99
7.1
ROTORUA
Rt3* (602)
Rt8*
Rtl2 (428)
Rtl3*(364)
Rtl4*(358)
Rtl5* (351)
Rt30*
Rt41 (360)
Rt42*
Rt51*(426)
Rt56*
Rt57*
Rt58 (711)
Rt59 (720)
Rt60(715)
Rt61
Rt68
Rt70
Rt71
Rt72
Rt74
Rt80
Rt82
Rt84
Rt87 (396)
Rt88 (392)
Rt89 (388)
Rt90 (386)
Rt91 (383)
Rt93
Rt98 (354)
Rt99 (363)
WAIMANGU
Wai 8
Wai9
Wai 11
Wai 17
Wai 18
Wai 19
Wai21
WAIRAKE1
Wkl
WAIKETI
VALLEY
Wkv6
Bacterial morphotype
Rods to filaments Cocci
METHODS
Enrichment and isolation
Pools covering the temperature range from 64 to
103°C and a pH range from 3.1 to 9.1 were sampled from a number of thermal regions of New
Zealand (Table 1). Procedures for sampling, pH,
and temperature measurement and for the preparation of anaerobic media were performed as previously described (Patel et al. 1985a, 1985b, 1985c).
Initial enrichments were performed in an anaerobic pool enrichment medium (PEM). PEM consisted of pool water as a basal medium to which
nutrient sources were added. An inoculum source
of 9 ml pool water and 1 ml sediment slurry was
added to sterile tubes. Other additions were made
to the tubes from sterile stock solutions: 0.1 ml of
1% trypticase peptone-yeast extract mixture, and
100 ul of 10% Na2S.9H,O. Twenty milligrams of
sulphur (non-sterilised) and resazurin were also
added to each tube. Air in the head space was
replaced with nitrogen by flushing with oxygen-free
nitrogen and the tube sealed with a Hungate cap.
Because of the reducing nature of the thermal sediments and the addition of sulphide, anaerobic conditions quickly prevailed as evidenced by resazurin
reduction. Replicate enrichments from each site
were incubated at 85, 92, and 95°C. Control tubes
were prepared similarly and in addition contained
0.4 ml of 40% formalin. These acted as controls for
determination of initial cell counts present before
enrichments. To determine cell growth in enrichments, initial and final cell numbers were determined using a PetrofF-Hausser cell counting
chamber (C. A. Hausser and Son, USA) as described
by the manufacturer.
Db medium was essentially as described by Zillig et al. (1981) except that the pH of the medium
was adjusted to 6.8. It was also used to initiate
enrichments and for all subsequent transfers of
cultures.
Isolation of pure cultures from enrichments was
not possible using conventional procedures involving agar because at the temperature of incubation
agars did not remain sufficiently gelled. For the rods,
pure cultures were obtained by repeated transfers
at 98°C until no cocci appeared when reinoculated
at 88°C. The culture was then taken through at least
two successive end-point serial dilutions. Enrich-
KETATAHI
KetlO
TOKAANU
Tokl2
TAUPO
Tp44
NA = Not analysed. Media used was as described in
Materials and Methods. An increase in cell numbers and
sulphur corrosion was used as an index of growth.
*Indicates that enrichments were performed at least twice
at different times and identical results were obtained.
Official DSIR numbers of the pools (wherever available)
have been listed in parenthesis.
441
Patel et al.—Archaebacteria from New Zealand hot springs
Table 2 Characteristics of Thermoproteus and Desulfurococcus species. ND = not determined.
Strain
COCCI
R159-S1
Tokl2-Sl
Desulfurococcus
mobilis
Desulfurococcus
mucosus
RODS
RH3-S1
Tokl2-S2
KetlO-Sl
Wkv6-Sl
Tp44-Sl
Thermoproteus
tenax
Thermoproteus
neutrophilus
Growth range
Temp (° Q p H
Growth optima
Temp (' Q p H
Generation
time (h)
Autotrophy
Growth on
casein
65-95
65-95
4.0-8.0
4.0-7.5
88
88
6.8
6.0
4.0
3.25
No
No
+
+
ND
4.5-7.0
85
6.0
3.0
No
+
ND
4.5-7.0
85
6.0
4.0
No
ND
75-98
75-98
75-98
70-98
82-98
4.5-7.5
4.5-7.5
4.5-8.0
4.5-8.0
5.0-8.0
90
6.0
90
6.5
5.5
8.0
90
88
90
6.5
6.4
6.9
13.5
No
No
No
6.5
6.0
Facultative
Facultative
ND
ND
88
5.5
1.7
Facultative
ND
ND
85
6.8
4-8
Obligate
In addition, D. mobilis but not D. mucosus is motile and D. mucosus but not D. mobilis has a slimy
polymer attached to its envelope.
ments with cocci were freed of contaminating rods
by successive transfer at 70°C on Db medium containing casein instead of peptone. When no rods
were present on subsequent incubations at 85°C on
normal Db medium, the culture was taken through
at least two successive end-point serial dilutions.
The pure cultures so obtained were stored by freeze
drying.
The numbers of organisms present in pools were
estimated from sediment slurry or pool water as
an inoculum source. The preparation of the sediment slurry for enumeration was as follows: Sediment as well as pool water was collected and the
sediment allowed to settle out for six hours, after
which the supernatant was siphoned out and a 1:1
dilution of the sediment prepared using the appropriate medium. A twenty fold dilution was made
by adding 1 ml of the sediment slurry to 9 ml of
medium, from which serial dilutions in Db medium
were prepared. When pool water was used, serial
dilutions were prepared directly without the twenty
fold dilution step. Triplicate tubes were inoculated
from each dilution and incubated at 88 and 95°C
for determination of bacterial numbers by the MPN
method (APHA 1973).
For nutrient utilisation tests, the yeast extract
concentration of Db medium was reduced to
0.0001% which was insufficient to sustain growth
of any cultures tested. Carbon sources were added
singly from filter sterilised stocks to a concentration of 0.2% or in combination with 0.02% yeast
extract. Either sulphur or 0.1% cystine was used as
electron acceptor. Sulphur was not sterilised before
use, and no growth was observed in uninoculated
control tubes which contained unsterilised sulphur.
Tubes were inoculated with 0.1 ml of inoculum to
avoid carry over of nutrients and incubated at 90°C.
After three days tubes were examined microscopically for growth. Cell numbers of tubes exhibiting
growth were counted using a Coulter Counter
(Model ZB1) with a lower threshold of 5, an aperture current of 1 mA, an amplitude of 1/2 (cocci)
or 1/4 (rods), and a counting tube of 30 \vm. orifice.
Mean values were corrected with the appropriate
coincidence factor [(mean value/1000)2 X 0.3375].
Where growth was observed the tubes were used
as an inoculum into the same medium to confirm
the growth and cell density obtained.
Growth and cellular characterisation
Inoculum
source
Electron microscopy and phase contrast microscopy were performed for cellular characterisation
as described previously (Patel et al. 1985a, 1985b,
1986).
Table 3 Most probable numbers of Thermoproteus
cells in pool Rtl3.
Pool water
Sediment slurry
No. of cells ml"1
C
95°C
1.1 X 103
4.3 X 104
4.6 X 102
9.3 X 10'
type
442
New Zealand Journal of Marine and Freshwater Research, 1986, Vol. 20
* : 'v.
*
*
LOjum
RESULTS AND DISCUSSION
In general, enrichments initiated in both media gave
rise to the same mixed bacterial population of rods,
filaments, and cocci at 85 and 95°C, whereas only
rods were present in enrichments at 98°C with the
exception of pool Rt42 where a coccus associated
with the rods was also present. The cell wall structure of the coccus in pool Rt42 differed from the
cocci enriched from other pools at lower incubation temperatures and was very similar to that of
a rod present in the pool and with which it was
frequently found in association (Fig. 1). A pure culture of the coccus was not obtained, and it was not
investigated further. Such cocci-rod associations
have been reported previously (Zillig et al. 1981).
Positive enrichments were obtained from 36 out
of the 44 pools lested (Table 1). Of the eight negative results, four of the pools had pH values below
3.7 which might explain the absence of neutrophilic organisms. However, two other springs also
below pH 3.7 (Rt71 and Wai9) gave positive
enrichments. Positive enrichments might have been
B
expected from the remaining four springs on the
basis of their pH and temperature profile, but
repeated attempts at enriching from springs Rt3 and
Rt8 were unsuccessful. This does not indicate that
such springs are sterile since both Rt3 and Rt8 possess a diverse flora of thermophilic eubacteria (Patel
1984; Patel et al. 1985b).
Seven pure cultures were obtained for further
investigation, five rods (Ket 10-S1, Rtl 3-S1, Tok 12S2, Wkv6-Sl, and Tp44-Sl) and two coccal (Rt59Sl and Tokl2-Sl). Salient properties of the cultures are compared with reported values for Thermoproteus and Desulfuwcoccus isolates from
Iceland (Table 2).
All the five rod shaped isolates resembled Thermoproteus tenax Krai in morphology (Zillig et al.
1981). They were 3-20 \xm by 0.4 urn in dimension
and formed terminal spherical protrusions, termed
golf balls, in exponential growth phase. During
approach of the stationary phase, constriction of
the cytoplasm occurred and refractile bodies, usually more than one per cell, were also observed.
Patel et al.—Archaebacteria from New Zealand hot springs
443
ft •
m
9
•
0 2jam
0
Fig. 1 An electron micrograph showing the association of a coccus and a rod-shaped archaebacterium observed in
an enrichment culture obtained from pool Rt42 (A). The coccus possessed flagella (B) and the cell envelope possessed
a distinct sub-unit structure (C) which appeared similar to that of the rod-shaped bacterium (D).
Electron microscopy of negatively stained preparations revealed that all the five isolates had cell
walls made of distinct subunits (similar to that
exemplified in Fig. ID) which were masked by slime
production in the late stationary phase of growth.
The temperature optima of the New Zealand isolates were between 88 and 90°C and similar to the
temperature optima of other Thermoproteus tenax
species (Fischer et al. 1983). However, the New
Zealand isolates grew optimally at pH 6.0 to 6.9;
this differs from Thermoproteus tenax (Zillig et al.
1981; Fischer et al. 1983) which grew optimally at
a pH of 5.5 with no growth observed at pH 7.0.
The New Zealand isolates also grew more slowly
with generation times between 5.5 and 13.5 h. These
features of the five isolates were similar to Thermoproteus neutrophilus (Fischer et al. 1983). However, T. neutrophilus is an obligate autotroph but
all five New Zealand isolates were either facultative autotrophs or strict heterotrophs (Table 2). This
is the first report on the isolation of strictly heterotrophic Thermoproteus species from geothermal
environments. It would be interesting to investigate whether they are unique to New Zealand
geothermal habitats.
No increase in cell numbers was obtained when
all the five New Zealand isolates and T. neutrophilus were grown on acetate, ethanol, lactose, glucose, pyruvate, or sucrose as carbon sources.
Additionally, T. tenax Krai in our hands was found
not to utilise glucose and ethanol as carbon sources
though Zillig et al. (1981) have reported that it does.
With the exception of the obligate autotroph T.
neutrophilus, all cultures showed good growth on
Db medium with yeast extract as sole carbon
source, i.e., no peptone, and an increase in cell
numbers was evident at yeast extract concentrations of only 0.01%. When different carbon sources
were added to Db medium the yeast extract must
be reduced to below 0.01% to avoid misinterpretation of the carbon source actually used. Malate
was the only substrate used as a sole carbon and
energy source by the New Zealand isolates and T.
tenax Krai but not by T. neutrophilus. Under opti-
444
New Zealand Journal of Marine and Freshwater Research, 1986, Vol. 20
sulphur or cystine as electron acceptor. This is in
contrast to observations by Zillig et al. (1982) on
sulphur free growth of D. mucosus. Yeast extract
did not support growth of either of the coccal
strains, both of which had an obligate requirement
for either peptone or casein.
All rod and coccal New Zealand isolates were
deemed to be archaebacteria as a consequence of
mode of reproduction (golf balls and budding
respectively), their anaerobic respiration of sulphur, morphology, and ultrastructure in relation to
type species and resistance to antibiotics particularly rifampicin and those inhibiting peptidoglycan
synthesis (Jasperse-Herst 1984).
The geographic distribution of thermophilic
archaebacteria is interesting. The Thermoproteus
isolates obtained from pools in Iceland and USA
have all been autotrophs or facultative autotrophs,
the three New Zealand strains are the only ones
Fig. 2 An electron micrograph of a Desulfurococcus species obtained from a New Zealand hot spring which reported incapable of autotrophic metabolism.
Desulfurococcus has only been isolated previously
reproduces by budding.
from Icelandic hot springs, and could not be isolated from pools in Italy or Japan. The New Zealand
pools would appear to be similar to Icelandic pools
mum growth conditions on Db medium all isolates in their archaebacterial flora, but with species
produced maximum cullure densities of over adapted to the more neutral pH of the former.
1X108 cells m l 1 after 48 h incubation.
Table 3 presents data on the estimation of numbers of Thermoproteus species present in pool Rtl 3
(sediment or pool water) at incubation tempera- ACKNOWLEDGMENT
tures of 88 and 95°C in Db medium. Slightly higher We gratefully acknowledge financial support from a BP
values were obtained at 88°C for both water and EMRA award to Mrs Jasperse-Herst.
sediment estimations than at 95°C. This was in
accordance with the known optimum temperature
for growth of both T. tenax (83°C) and for pure
cultures (88-90°C) from New Zealand thermal pools REFERENCES
(Table 2).
APHA. Water Pollution Control Federation 1973: StandThe greater number of cells found in pool sediard methods for the examination of water and
wastewater, 13th ed. Washington, pp. 672-676.
ments might indicate a lower Eh and/or higher
substrate concentration in this environment. If the Brock. T. D. 1978: Life in boiling water. In: Thermorods were adsorbed to sediment particles then the
philic microorganisms and life at high temperatures. New York, Berlin, Heidelberg, Springer
MPN approach might significantly underestimate
Verlag. pp. 302-336.
the number of cells in the sediment, because the
Buonocore,
V.; Sgambati, O.; De Rosa, M.; Esposito, E.;
number of cells per particle could be high.
Gambacorta, A. 1980: A constitutive fi-galactosidThe New Zealand coccal strains closely resemble
ase from the extreme thermoacidophile archaeDesulfurococcus species. The New Zealand coccal
bacterium Caldariella acidophila : properties of the
isolates were Gram-negative, non-motile, and varenzyme in the free state and in immobilized whole
ied in diameter from 2.0-7.0 \xm. Reproduction was
cells. Journal of applied biochemistry 2 : 390-397.
by budding (Fig. 2). In late stationary phase an Fischer, F.; Zillig, W.; Stetter, K. O.; Schreiber, G. S. 1983:
increasing number of cells appeared as membrane
Chemolithotrophic metabolism of anaerobic
extremely thermophilic archaebacteria. Nature 301:
ghosts. Motility is a feature of D. mobilis but not
511-513.
D. mucosus and hence the: New Zealand isolates
appeared to be similar to D. mucosus rather than Fox, G. E.; Stackebrandt, E.; Hespell, R. B.; Gibson, J.;
ManilofF, D.; Dyer, T. A.; Wolfe, R. S.; Balch, W.
D. mobilis. However, both New Zealand isolates
E.; Tanner, R. S.; Magrum, L. J.; Zablen, L. B.;
had slightly higher temperature optima than D.
Blakemore, R.; Gupta, R.; Bonen, L; Lewis, B. J.;
mucosus and isolate Rt59-Sl had a significantly difStahl, D. A.; Luehrsen, K. R.; Chen, K. N.; Woese,
ferent pH optimum (Table 2). Both New Zealand
C. R. 1980: The phylogeny of prokaryotes. Science
strains were unable to grow in media without either
209: 457-463.
Patel et al.—Archaebacteria from New Zealand hot springs
Jasperse-Herst, P. M. 1984: Isolation and characterization of extremely thermophilic archaebacteria from
New Zealand hot springs. Unpublished M Phil
thesis, University of Waikato, Hamilton, New
Zealand.
Patel, B. K. C. 1984: Extremely thermophilic bacteria in
New Zealand hot springs. Unpublished PhD thesis, University of Waikato, Hamilton, New
Zealand.
Patel, B. K. C; Morgan, H. W.; Daniel, R. M. 1985a:
Fervidobacterium nodosum gen. nov. and sp. nov.,
a new chemo-organotrophic, caldoactive, anaerobic bacterium. Archives of microbiology 141 : 6369.
1985b: Thermophilic anaerobic spirochetes in
New Zealand hot springs. FEMS microbiology letters 26: 101-106.
1985c: A simple and efficient method for preparing and dispensing anaerobic media. Biotechnology letters 7: 277-278.
1986: Unusual microorganisms observed in
New Zealand hot springs. Microbial ecology (in
press).
Stetter, K. O.; Thomm, M.; Winter, J.; Wildgruber, G.;
Huber, H.; Zillig, W.; Janekovic, D.; Konig, H.;
Palm, P.; Wunderl, S. 1981: Methanothermus fervidus sp. nov., a novel extremely thermophilic
methanogen isolated from an Icelandic hot spring.
Zentralblatt fur bakteriologie und hygiene. I. Abteilung originale, C2: 166-178.
445
Stetter, K. O.; Konig, H.; Stackebrandt, E. 1983: Pyrodictium gen. nov., a new genus of submarine discshaped sulphur reducing archaebacteria growing
optimally at 105°C. Systematic and applied microbiology 4: 535-551.
Zillig, W.; Stetter, K. O.; Schafer, W.; Janekovic, D.;
Wunderl, S.; Holz, I.; Palm, P. 1981: Thermoproteales: A novel type of extremely thermoacidophilic anaerobic archaebacteria isolated from
Icelandic solfataras. Zentralblatt filr bakteriologie
und hygiene I, Abteilung originale C2 : 205-227.
Zillig, W.; Stetter, K. O.; Prangishvilli, D.; Schafer, W.;
Wunderl, S.; Janekovic, D.; Holz, I.; Palm, P. 1982:
Desulfurococcaceae, the second family of the
extremely thermophilic anaerobic sulphur-respiring Thermoproteales. Zentralblatt fur bakteriologie und hygiene I. Abteilung originale C3: 304317.
Zillig, W.; Gierl, A.; Schreiber, G.; Wunderl, S.; Janekovic, D.; Stetter, K. O.; Klenk, H. P. 1983a: The
archaebacterium Thermofilum pendens represents
a novel genus of the thermophilic, anaerobic sulphur respiring Thermoproteales. Systematic and
applied microbiology 4 : 79-83.
Zillig, W.; Holz, I.; Janekovic, D.; Schaefer, W.; Reiter,
W. D. 1983b: The archaebacterium Thermococcus
celer represents a novel genus within the thermophilic branch of archaebacteria. Systematic and
applied microbiology 4: 88-94.